CN114358165A - Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion - Google Patents
Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion Download PDFInfo
- Publication number
- CN114358165A CN114358165A CN202111624830.8A CN202111624830A CN114358165A CN 114358165 A CN114358165 A CN 114358165A CN 202111624830 A CN202111624830 A CN 202111624830A CN 114358165 A CN114358165 A CN 114358165A
- Authority
- CN
- China
- Prior art keywords
- data
- photovoltaic module
- ground fault
- source
- weather forecast
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Photovoltaic Devices (AREA)
Abstract
The invention discloses a detection method for preventing a photovoltaic module ground fault based on multi-source data fusion, which comprises the following steps: acquiring data sources of different space-time dimensions of the photovoltaic module; selecting different data analysis methods to analyze the data source states of the corresponding types; and fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module. According to the method, the multi-dimensional data sources are subjected to fusion analysis, so that the earth fault rate of the photovoltaic module is accurately predicted, the earth fault of the photovoltaic module is effectively prevented, and the continuous and stable power generation of the photovoltaic module is guaranteed.
Description
Technical Field
The invention relates to the technical field of photovoltaic power generation, in particular to a detection method for preventing a photovoltaic module ground fault based on multi-source data fusion.
Background
The photovoltaic module is used as core equipment of a photovoltaic power generation system, the working efficiency of the photovoltaic module directly influences the economic benefit of an enterprise, and fault treatment and daily maintenance of the photovoltaic module are important measures for ensuring the availability of the power generation core equipment. Increasing problems and accidents in the field indicate that a ground fault of a photovoltaic module is one of the most common faults of a photovoltaic power plant.
At present, a detection method for detecting the ground fault of a photovoltaic module is mainly carried out by adopting a detection circuit, and when the detection is carried out, the ground fault of the photovoltaic module occurs, so that power failure loss is caused.
Therefore, how to provide a detection method capable of preventing the ground fault of the photovoltaic module is a problem that needs to be solved urgently by those skilled in the art.
Disclosure of Invention
In view of the above, the invention provides a detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion, which realizes accurate prediction of the ground fault rate of the photovoltaic module, effectively prevents the ground fault of the photovoltaic module and ensures continuous and stable power generation of the photovoltaic module by performing fusion analysis on a multi-dimensional data source.
In order to achieve the purpose, the invention adopts the following technical scheme:
a detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion comprises the following steps:
acquiring data sources of different space-time dimensions of the photovoltaic module;
selecting different data analysis methods to analyze the data source states of the corresponding types;
and fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module.
Preferably, in the above detection method for preventing ground fault of a photovoltaic module based on multi-source data fusion, the method further includes:
and inquiring corresponding measures from a pre-constructed knowledge base and distributing the measures to corresponding maintenance terminals based on the ground fault probability of the photovoltaic module and the current position of the photovoltaic module.
Preferably, in the above detection method for preventing ground fault of a photovoltaic module based on multi-source data fusion, the method further includes:
and writing the data source state, the ground fault probability, the actual ground fault and the corresponding measures of the photovoltaic module into the knowledge base, and updating the knowledge base.
Optionally, in the above detection method for preventing ground fault of a photovoltaic module based on multi-source data fusion, the type of the data source at least includes: video imaging data, family defect data, online operation parameters, operation maintenance data, vibration sensor data, weather forecast data and work ticket data of a photovoltaic power station of the photovoltaic module.
Optionally, in the above detection method for preventing the ground fault of the photovoltaic module based on the multi-source data fusion, the video imaging data of the photovoltaic module is obtained by an unmanned aerial vehicle or a low-balloon mounted camera; the family defect data is obtained from a pre-constructed knowledge base; the weather forecast data is acquired from a server interface; and the online operation parameters, the operation maintenance data and the vibration sensor data are acquired from a photovoltaic operation and maintenance management system.
Optionally, in the above detection method for preventing the ground fault of the photovoltaic module based on the multi-source data fusion, an image search matching algorithm is adopted to analyze video imaging data of the photovoltaic module; analyzing the family defect data by adopting a fuzzy mathematics method; analyzing short-time short-term weather forecast data in the weather forecast data by adopting an LSTM algorithm, analyzing medium-term weather forecast data in the weather forecast data by adopting a BP neural network algorithm, and performing weighted average on analysis results of the short-time weather forecast data, the short-term weather forecast data and the medium-term weather forecast data; analyzing the online operation parameters, the operation maintenance data and the vibration sensor data by adopting a statistical method; and analyzing the work ticket data of the photovoltaic power station by adopting a query or search method.
Optionally, in the above detection method for preventing the ground fault of the photovoltaic module based on the multi-source data fusion, a process of analyzing the video imaging data of the photovoltaic module includes:
extracting a video frame in the video data of the photovoltaic module, comparing the video frame with the original state image of the photovoltaic module area by adopting an image search matching algorithm, and judging the consistency degree between the video frame and the original state image of the photovoltaic module area;
marking the video frames with the consistency degree lower than a preset value as warning images;
and identifying the surface state of the photovoltaic module in the warning image, and judging whether field construction operation exists around the warning image.
Optionally, in the above detection method for preventing ground fault of a photovoltaic module based on multi-source data fusion, the fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module includes:
carrying out weighted average on state analysis results of different data sources by adopting a data fusion algorithm to obtain a calculation result of the fault information of the photovoltaic module; the calculation result comprises a fault type, a fault position and a ground fault probability value;
fitting and verifying the calculation result of the data fusion algorithm by adopting a BP neural network algorithm, and predicting the probability value of the ground fault of the output photovoltaic module;
and comparing the ground fault probability value in the calculation result of the data fusion algorithm with the ground fault probability value output by the BP neural network algorithm, wherein if the deviation value between the two values is within a preset value, the current calculation result is valid, and if the deviation value between the two values is not within the preset value, the current calculation result is invalid.
Optionally, in the above detection method for preventing the ground fault of the photovoltaic module based on the multi-source data fusion, an expression of the data fusion algorithm is as follows:
Y1＝∑yibi；
where i denotes the data source type, yiRepresenting the results of the state analysis of the ith class of data sources, biRepresents a weight of a state analysis result of the i-th class data source, and 1 ═ Σ bi。
According to the technical scheme, compared with the prior art, the method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion is used for carrying out data fusion on video imaging data of the photovoltaic module, short-term and medium-term data of weather forecast, family information of the photovoltaic module, operation and maintenance data, online operation parameters of the photovoltaic module, vibration sensor data and a work ticket of a photovoltaic power station, calculating the probability of the ground fault of the photovoltaic module, and automatically generating a countermeasure after the ground fault of the photovoltaic module, so that the ground fault of the photovoltaic module is effectively prevented, and continuous and stable power generation of the photovoltaic module is guaranteed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion according to an embodiment of the present invention;
fig. 2 is a flowchart of a detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion in another embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the invention discloses a detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion, which includes the following steps:
s1, acquiring data sources of different space-time dimensions of the photovoltaic module;
s2, selecting different data analysis methods to analyze the data source states of the corresponding types;
and S3, fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module.
In one embodiment, further comprising:
and S4, based on the ground fault probability of the photovoltaic module and the current position of the photovoltaic module, inquiring corresponding measures from a pre-constructed knowledge base and distributing the measures to corresponding maintenance terminals. Wherein the family defect information of the photovoltaic module is obtained from a knowledge base. And when the ground fault probability exceeds a set threshold value, judging that a fault exists, and inquiring a countermeasure from the knowledge base.
In other embodiments, further comprising:
as shown in fig. 2, S5, writing the data source status, the ground fault probability, the actual ground fault and the countermeasure of the photovoltaic module into the knowledge base, and updating the knowledge base.
The actual ground fault is obtained from the actual record. After each fault occurs, the actually occurring fault type, time, processing measures and other related contents are recorded in the knowledge base for updating the knowledge base. Strictly speaking, each fault is different, the occurrence time is different, the conditions are different, and the actually occurring fault is added into the knowledge base, so that the capability of the knowledge base is enhanced
The embodiment of the invention can also automatically set the weight, and the weight is obtained according to the photovoltaic module knowledge base.
According to the embodiment of the invention, different setting methods are adopted to automatically set the weight according to the target required to be set; for example, when the weight is set when the weather condition tends to be emphasized, the weight value of the weather condition is increased.
The above steps are further described below.
In S1, the types of the data source at least include: video imaging data, family defect data, online operation parameters, operation maintenance data, vibration sensor data, weather forecast data and work ticket data of a photovoltaic power station of the photovoltaic module.
The time-related data sources include family defect data of photovoltaic components, operation maintenance data, short-term and medium-term weather forecast data and work tickets of photovoltaic power stations. The data sources relating to the space are video imaging data of the photovoltaic modules, vibration sensor data and work tickets of the photovoltaic power station. The basic data source of the photovoltaic module is the operation parameters of the photovoltaic module, such as voltage, current, heat, position information and self position information.
The family defect data refers to the defect data of the same manufacturer of the photovoltaic module; different manufacturers are different families. In practical applications, if a photovoltaic module of a certain manufacturer has defects, the product of the photovoltaic module may have the defects generally. By using this characteristic, a higher weight value can be given to the index.
The time, the place, the area range, the working category and the content of the construction, the maintenance and other work of the photovoltaic power station are recorded on the work ticket of the photovoltaic power station. If there is near the photovoltaic module operation such as digging pit, demolish, installation, then destroy probability greatly increased to the photovoltaic module, for example often there is the construction to dig the condition of breaking the cable, digging bad photovoltaic module basis.
The data sources are obtained as follows:
and video imaging data of the photovoltaic module is collected by a high-altitude camera. Specifically, unmanned aerial vehicle or low balloon carry camera through wireless 4G and 5G signal, transmits to video acquisition service device earlier, is transmitted to the monitoring service platform through optic fibre by video acquisition service device again and carries out video analysis and data processing, discerns photovoltaic module surface state, judges simultaneously whether there is the site operation. The video acquisition service device is responsible for collecting and storing video image data of the camera and performing some preliminary processing; the monitoring service platform is responsible for monitoring the operation state of a photovoltaic assembly of a photovoltaic power station, video data and collected data of a sensor, such as voltage, current, power generation and the like are provided, the data volume of video image data is large, and the data of the voltage, the current, the power generation and the like of the photovoltaic assembly are large in number, but the storage space occupied by each point is small, and each point only occupies 4 bytes.
Family defect data is obtained from a pre-constructed photovoltaic module knowledge base.
The weather forecast data of short-term, short-term and medium-term periods is acquired from the server interface.
And acquiring the operation maintenance data, the operation parameters and the vibration sensor data from the photovoltaic operation and maintenance management system.
And the work order of the photovoltaic power station is issued by the dispatching system and is acquired from the photovoltaic operation and maintenance management system.
In S2, the specific process of selecting different data analysis methods to analyze the data source states of the corresponding types is as follows:
and analyzing the video imaging data of the photovoltaic module by adopting an image search matching algorithm. The method specifically comprises the following steps:
1. extracting a video frame in the video data of the photovoltaic module, comparing the video frame with the original state image of the photovoltaic module area by adopting an image search matching algorithm, and judging the consistency degree between the video frame and the original state image of the photovoltaic module area;
2. marking the video frames with the consistency degree lower than a preset value as warning images; if the consistency degree is more than 95%, the image is an effective image; and if the consistency degree is lower than 95%, setting an alarm mark as an alarm image based on an image obtained by the camera video mounted on the low-air ball.
3. And identifying the surface state of the photovoltaic module in the warning image, and judging whether field construction operation exists around the warning image. By adopting an image identification method, abnormal states of the photovoltaic module, such as missing, overturning, breaking and the like, are identified, meanwhile, construction operation sites around the photovoltaic module are analyzed, and key areas are set.
Analyzing the family defect data by adopting a fuzzy mathematics method; the product quality is divided into three grades of good, general and poor, and the corresponding weight is 2, 1 and 0.
Analyzing short-term and short-term weather forecast data in the weather forecast data by adopting an LSTM algorithm, analyzing medium-term weather forecast data in the weather forecast data by adopting a BP neural network algorithm, and performing weighted average on analysis results of the short-term, short-term and medium-term weather forecast data, wherein the weights are 1/3 respectively.
Analyzing the online operation parameters, the operation maintenance data and the vibration sensor data by adopting a statistical method; and calculating a maximum value, a minimum value, a mean square error and the like, and identifying whether the data have abnormality or not.
The position of the photovoltaic module is obtained through analyzing the work ticket of the photovoltaic power station, the influence degree of the photovoltaic module at the position is 1, and the influence degrees of the photovoltaic modules at other positions are 0. . The current work ticket can also form records in the system when printing paper; the work ticket contains the work content, time, personnel, place, measures and the like. Therefore, the position information of the photovoltaic module can be easily obtained from the work ticket by adopting a query or search method.
In S3, fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module, which specifically includes:
carrying out weighted average on state analysis results of different data sources by adopting a data fusion algorithm to obtain a calculation result of the fault information of the photovoltaic module; the calculation result comprises a fault type, a fault position and a ground fault probability;
fitting and verifying the calculation result of the data fusion algorithm by adopting a BP neural network algorithm, and predicting the probability value of the ground fault of the output photovoltaic module;
and comparing the ground fault probability value in the calculation result of the data fusion algorithm with the ground fault probability value output by the BP neural network algorithm, if the deviation value between the two values is within a preset value, the current calculation result is valid, and if the deviation value between the two values is not within the preset value, the current calculation result is invalid.
Specifically, the expression of the data fusion algorithm is as follows:
Y1＝∑yibi；
where i denotes the data source type, yiRepresenting the results of the state analysis of the ith class of data sources, biRepresents a weight of a state analysis result of the i-th class data source, and 1 ═ Σ bi。
In one embodiment, yiAnd (i is 1, 2, 3, 4, 5 and 6) which are respectively a video analysis result, a short-time short-term and medium-term weather forecast data processing result, a photovoltaic module online state parameter processing result, a photovoltaic module vibration sensor data processing result, a photovoltaic power station work ticket processing result and a photovoltaic module family defect information processing result.
biAnd (i is 1, 2, 3, 4, 5 and 6) which are respectively the weight of the video analysis result, the short-term and medium-term weather forecast result, the photovoltaic module online state parameter processing result, the photovoltaic module vibration sensor processing result, the photovoltaic power station work ticket processing result and the photovoltaic module family defect information processing result.
The processing results of the six data sources Yi are obtained by processing different algorithms of different data sources. The method comprises the following steps:
the video analysis result is obtained by processing video image data of a camera mounted on an unmanned aerial vehicle or a low-air balloon, and the content of the video analysis result comprises the following steps: data source type, photovoltaic module fault type, photovoltaic module position, probability of fault occurrence.
The short-term, short-term and medium-term weather forecast data processing result is obtained by calculating the weight of future influence according to the acquired weather forecast data, and the weight value can increase the contribution values of other calculation results. When the weather deteriorates in the future (i.e. the non-illumination time is long), the weight value is large; when the weather is clear in the future, the weight value is small and approaches to 0. By adopting a fuzzy mathematics method, the future weather conditions are respectively corresponding to weighted values {0.01, 0.2, 0.6, 0.8, 1.5 and 1.8} from good to bad, namely { very good, general, poor, very poor and particularly bad }. Very good, namely in the future two weeks, the weather is clear and the illumination is sufficient. Good, namely, the weather is clear in the future one week, the illumination is sufficient, but the possibility of cloudiness exists, and the weather is uncertain after one week. Generally, the weather is better in the next two days, and there may be a cloud condition, and the weather after two days has a cloud or cloudy days. Worse, a cloudy day within three days in the future. Very poor, that is, no light exists in the future one week, and strong wind, precipitation and the like exist. Particularly bad, namely, no light is emitted in the future two weeks, and rain, snow, hail, strong wind and the like exist in the weather. These weather conditions correspond to the above weight values.
And judging the state of the photovoltaic module, including the position of the photovoltaic module and the state of the photovoltaic module, according to the monitored parameters of the voltage, the current, the generating capacity and the like of the photovoltaic module on-line state parameter processing result. The state of the photovoltaic module is divided into { very good, general, bad and very bad } according to a fuzzy mathematic method, and the corresponding numerical values are {1.0, 0.8, 0.5, 0.2 and 0.1 }.
The data processing result of the vibration sensor of the photovoltaic module is that the states are distinguished according to the online data of the vibration sensor, and when the data of the vibration sensor is higher than a specified value, the state is a bad state; when the data of the vibration sensor is satisfactory, it is in a good state. The corresponding data value is 0.2, 1.0. And the data processing result also comprises the position information of the photovoltaic module.
The mode of the photovoltaic power station work ticket processing result and the photovoltaic module family defect information processing result is similar to the mode of the photovoltaic power station work ticket processing result and the photovoltaic module family defect information processing result.
The weight bi distribution principle of the above six data sources is as follows:
sorting from big to small: the method comprises the steps of processing results of photovoltaic module online state parameters, processing results of a photovoltaic module vibration sensor, video analysis results, processing results of a photovoltaic power station work ticket, short-term and short-term weather forecast results and processing results of photovoltaic module family defect information. Their weighted values add up to 1. For example, values {0.30, 0.25, 0.20, 0.15, 0.08, 0.02} may be taken. 0.30+0.25+0.20+0.15+0.08+0.02 ═ 1.0.
The specific process of verifying the first earth fault probability value by adopting the BP neural network algorithm comprises the following steps:
the actual result and the calculation result of the data fusion algorithm are stored in each calculation; wherein, the actual result is stored by the calculation result after the fault occurs; when a failure does not occur, it is stored as 0. Thus, after the accumulated data is more, the data becomes the training data of the BP neural network algorithm.
Predicting new data by the BP neural network algorithm according to the accumulated data; and comparing the new data with the calculation result of the data fusion algorithm, and observing the difference level of the new data. If the difference is not significant, the results are valid. If the difference is significant, the results are invalid.
And carrying out BP neural network training on the calculation result of the data fusion algorithm, then predicting the future result, and comparing and verifying the predicted result with the result of the multi-source data fusion analysis first algorithm. In general, there will be some deviation; if the deviation value is large and the relative value exceeds 30%, a problem exists, and the current calculation result is invalid; if the deviation is small and within 10%, the current calculation result is valid.
The result output by the data fusion algorithm is fault information of the photovoltaic module, including fault type, fault position and fault probability; of importance is a probability value. The essence of the method is to obtain the classified probability value to determine the maximum probability of the fault.
When the BP neural network carries out prediction, the output result of the data fusion algorithm is related, and the prediction is carried out by using the calculation result of the data fusion algorithm. And comparing the result after prediction with the result of the data fusion algorithm. That is, the data fusion algorithm is similar to the coarse screening process, and the BP neural network algorithm is similar to reprocessing once. Since the BP neural network algorithm is a non-linear algorithm, the results of the two times are not completely consistent. When the deviation is small, the prediction result is considered to be reliable. When the deviation is large, the prediction result is not credible, the failure cannot be judged, and the reason needs to be further analyzed.
In S4, the position information, fault type, fault description, fault generation reason, fault countermeasure and other records of the photovoltaic module are recorded in a knowledge base; according to the probability of the ground fault of the photovoltaic module, the fault type and the fault description are determined, and the corresponding measures are found by combining the position information of the photovoltaic module. Because of the same kind of failures at different locations, the countermeasures are different. And the task automatic distribution is to form a task list and an operation guide for the countermeasure of the photovoltaic fault, distribute the task list and the operation guide to the corresponding maintenance terminal and process the task list and the operation guide by maintenance personnel.
In S5, comparing the ground fault probability of the photovoltaic module with the actual fault, writing the data source state, the actual ground fault, and the corresponding measure of the photovoltaic module corresponding to the ground fault probability of the photovoltaic module into the knowledge base, and updating the knowledge base. The purpose is to generate valuable knowledge information and provide or retrieve information for the following new photovoltaic module state and the occurrence of the photovoltaic module ground fault.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A detection method for preventing a ground fault of a photovoltaic module based on multi-source data fusion is characterized by comprising the following steps:
acquiring data sources of different space-time dimensions of the photovoltaic module;
selecting different data analysis methods to analyze the data source states of the corresponding types;
and fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module.
2. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion is characterized by further comprising the following steps:
and inquiring corresponding measures from a pre-constructed knowledge base and distributing the measures to corresponding maintenance terminals based on the ground fault probability of the photovoltaic module and the current position of the photovoltaic module.
3. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion is characterized by further comprising the following steps:
and comparing the ground fault probability of the photovoltaic module with the actual fault, writing the data source state, the actual ground fault and the corresponding measure of the photovoltaic module corresponding to the ground fault probability of the photovoltaic module into the knowledge base, and updating the knowledge base.
4. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion as claimed in claim 2, wherein the types of the data sources at least comprise: video imaging data, family defect data, online operation parameters, operation maintenance data, vibration sensor data, weather forecast data and work ticket data of a photovoltaic power station of the photovoltaic module.
5. The multi-source data fusion-based detection method for preventing the ground fault of the photovoltaic module according to claim 4, wherein video imaging data of the photovoltaic module is obtained by an unmanned aerial vehicle or a low-air-ball mounting camera; the family defect data is obtained from a pre-constructed knowledge base; the weather forecast data is acquired from a server interface; and the online operation parameters, the operation maintenance data and the vibration sensor data are acquired from a photovoltaic operation and maintenance management system.
6. The multi-source data fusion-based detection method for preventing the ground fault of the photovoltaic module according to claim 4, characterized in that an image search matching algorithm is adopted to analyze video imaging data of the photovoltaic module; analyzing the family defect data by adopting a fuzzy mathematics method; analyzing short-time short-term weather forecast data in the weather forecast data by adopting an LSTM algorithm, analyzing medium-term weather forecast data in the weather forecast data by adopting a BP neural network algorithm, and performing weighted average on analysis results of the short-time weather forecast data, the short-term weather forecast data and the medium-term weather forecast data; analyzing the online operation parameters, the operation maintenance data and the vibration sensor data by adopting a statistical method; and analyzing the work ticket data of the photovoltaic power station by adopting a query or search method.
7. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion is characterized in that the process of analyzing the video imaging data of the photovoltaic module comprises the following steps:
extracting a video frame in the video data of the photovoltaic module, comparing the video frame with the original state image of the photovoltaic module area by adopting an image search matching algorithm, and judging the consistency degree between the video frame and the original state image of the photovoltaic module area;
marking the video frames with the consistency degree lower than a preset value as warning images;
and identifying the surface state of the photovoltaic module in the warning image, and judging whether field construction operation exists around the warning image.
8. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion of claim 1, wherein the fusing and verifying the state analysis results of the data sources to obtain the ground fault probability of the photovoltaic module comprises:
carrying out weighted average on state analysis results of different data sources by adopting a data fusion algorithm to obtain a calculation result of the fault information of the photovoltaic module; the calculation result comprises a fault type, a fault position and a ground fault probability value;
fitting and verifying the calculation result of the data fusion algorithm by adopting a BP neural network algorithm, and predicting the probability value of the ground fault of the output photovoltaic module;
and comparing the ground fault probability value in the calculation result of the data fusion algorithm with the ground fault probability value output by the BP neural network algorithm, wherein if the deviation value between the two values is within a preset value, the current calculation result is valid, and if the deviation value between the two values is not within the preset value, the current calculation result is invalid.
9. The method for detecting the ground fault of the photovoltaic module based on the multi-source data fusion is characterized in that the expression of the data fusion algorithm is as follows:
Y1＝∑yibi；
where i denotes the data source type, yiRepresenting the results of the state analysis of the ith class of data sources, biRepresents a weight of a state analysis result of the i-th class data source, and 1 ═ Σ bi。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111624830.8A CN114358165A (en) | 2021-12-28 | 2021-12-28 | Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111624830.8A CN114358165A (en) | 2021-12-28 | 2021-12-28 | Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114358165A true CN114358165A (en) | 2022-04-15 |
Family
ID=81104142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111624830.8A Pending CN114358165A (en) | 2021-12-28 | 2021-12-28 | Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114358165A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115965571A (en) * | 2022-04-28 | 2023-04-14 | 锋睿领创(珠海)科技有限公司 | Multi-source information fusion detection and model training method and medium for incremental autonomous learning |
-
2021
- 2021-12-28 CN CN202111624830.8A patent/CN114358165A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115965571A (en) * | 2022-04-28 | 2023-04-14 | 锋睿领创(珠海)科技有限公司 | Multi-source information fusion detection and model training method and medium for incremental autonomous learning |
| CN115965571B (en) * | 2022-04-28 | 2023-08-22 | 锋睿领创(珠海)科技有限公司 | Multi-source information fusion detection and model training method and medium for incremental autonomous learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111555716B (en) | Method, device, equipment and storage medium for determining working state of photovoltaic array | |
| CN111539550A (en) | Method, device and equipment for determining working state of photovoltaic array and storage medium | |
| CN115395646B (en) | Intelligent operation and maintenance system of digital twin traction substation | |
| CN112886923B (en) | Photovoltaic power station operation and maintenance method and device in thunder and lightning weather | |
| CN115329812A (en) | Road infrastructure abnormity monitoring method based on artificial intelligence | |
| CN117390403B (en) | Power grid fault detection method and system for new energy lighthouse power station | |
| CN117078017A (en) | Intelligent decision analysis system for monitoring power grid equipment | |
| CN115331403B (en) | Visual analysis method and system for fault data of power supply line | |
| CN114693146A (en) | Natural protection area foreign invasive species risk early warning system and method | |
| CN110690699A (en) | Transformer substation intelligent detection system based on ubiquitous power Internet of things | |
| Song et al. | Deformable YOLOX: Detection and rust warning method of transmission line connection fittings based on image processing technology | |
| CN114358165A (en) | Detection method for preventing ground fault of photovoltaic module based on multi-source data fusion | |
| CN113011477B (en) | Cleaning and completing system and method for solar irradiation data | |
| CN117494009A (en) | Electrical equipment state evaluation method based on insulating material pyrolysis analysis and cloud platform | |
| CN112016739B (en) | Fault detection method and device, electronic equipment and storage medium | |
| CN117949887A (en) | Fault detection and prevention method for electric energy metering system of energy controller | |
| CN117614137A (en) | Power distribution network optimization system based on multi-source data fusion | |
| CN116243103A (en) | Power transmission and transformation equipment state intelligent sensing and early warning method based on multi-source data | |
| CN115664006A (en) | Increment distribution network intelligence management and control integration platform | |
| CN115187880A (en) | Communication optical cable defect detection method and system based on image recognition and storage medium | |
| CN113992151A (en) | Method, device and equipment for determining working state of photovoltaic array and storage medium | |
| Segovia et al. | Connected system for monitoring electrical power transformers using thermal imaging | |
| CN111489077A (en) | Reliability assessment method for power communication system | |
| CN117726959B (en) | Unmanned aerial vehicle power line safety inspection system and method based on intelligent image recognition | |
| CN117197770B (en) | Inspection complete flow data supervision system and method based on Internet of things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |